Dimethyl fumarate has been a mainstay of therapy for relapsing-remitting multiple sclerosis for years, valued for its combination of efficacy, oral administration, and a generally favorable long-term safety profile. The drug exerts immunomodulatory effects through several mechanisms, including activation of the nuclear factor erythroid 2-related factor 2 pathway and a characteristic reduction in circulating lymphocytes. That lymphocyte decline is usually mild to moderate and reversible, but it is precisely the reason that treatment guidelines mandate regular white blood cell counting. The Belgian authors emphasize that patient-specific factors can modulate the leukocyte response to the drug, a point supported by prior research showing that baseline counts, body characteristics, and other variables influence how far lymphocytes fall in individual patients.

The clinical sequence at the heart of the report follows a pattern that respiratory physicians know well: influenza first, bacterial pneumonia second. Influenza A damages the respiratory epithelium, disrupts mucociliary clearance, and creates conditions in which colonizing bacteria such as Streptococcus pneumoniae can invade the lower airways. In an immunocompetent host this secondary bacterial pneumonia can still become life-threatening; in a patient with drug-induced lymphopenia the barrier to invasion is substantially lower. The patient in this report progressed to severe, invasive disease requiring intensive care, illustrating how a common seasonal virus and an opportunistic bacterium can combine catastrophically when immune defenses are compromised.

Severe pneumococcal pneumonia of this kind is not merely a chest infection that responds slowly to antibiotics. Community-acquired pneumonia that deteriorates into acute respiratory distress syndrome carries substantial mortality even in modern intensive care units, and studies of mechanically ventilated patients with pneumonia have documented how frequently this syndrome complicates the course. The global definition of acute respiratory distress syndrome was updated in 2024 to improve recognition of this entity, which involves diffuse lung inflammation, impaired oxygenation, and bilateral radiographic opacities. The Belgian team situates their case within this framework, noting that the combination of viral priming, bacterial invasion, and impaired cellular immunity created the conditions for profound respiratory failure.

The first of the report’s two central lessons concerns vaccination. People with multiple sclerosis are not inherently excluded from standard vaccine recommendations; on the contrary, the ECTRIMS/EAN consensus published in 2023 provides detailed guidance on immunization in this population, urging clinicians to complete recommended vaccines before starting highly active immunotherapies and to maintain coverage thereafter. Evidence indicates that immune responses to vaccines are generally maintained in patients treated with dimethyl fumarate, and a prospective single-center study in 2024 found that seasonal influenza and pneumococcal polysaccharide vaccines remain immunogenic in patients on various disease-modifying therapies. Yet a multicenter cross-sectional study published in 2025 documented that vaccination coverage among patients with multiple sclerosis remains incomplete and variable, with determinants ranging from disease characteristics to health system factors. The case therefore functions as a concrete reminder that every patient starting or continuing immunotherapy should have a documented vaccination history, with influenza and pneumococcal vaccines addressed explicitly.

The second lesson concerns hematologic monitoring. Treatment labels and consensus recommendations call for lymphocyte counting before initiation, at three and six months, and periodically thereafter, with dose adjustment or interruption when counts fall below defined thresholds. The Common Terminology Criteria for Adverse Events provides the grading framework used to classify the severity of leukopenia and lymphopenia, and the authors apply this vocabulary to describe the degree of immune suppression in their patient. Research into interventions to increase leukocyte testing during dimethyl fumarate treatment has shown that adherence to monitoring schedules is imperfect in real-world practice, and that structured interventions can improve testing rates. The report argues that the effort required for a simple blood test is small compared with the cost of missing a progressive lymphopenia that leaves a patient vulnerable to invasive bacterial disease.

The interplay between the two lessons deserves emphasis. Vaccination reduces the probability of the triggering infections, while monitoring detects the immune vulnerability that turns those infections into emergencies. Neither measure alone is sufficient. A fully vaccinated patient can still encounter a pneumococcal serotype not covered by the vaccine or an influenza strain drifted beyond the seasonal match, and a patient with normal lymphocyte counts can still suffer severe pneumonia. But the two safeguards together shrink the risk in complementary ways: vaccination lowers exposure to the pathogens most likely to cause invasive disease, and monitoring ensures that the pharmacologically induced component of susceptibility is caught and corrected before a routine viral illness escalates. German consensus recommendations on vaccination strategies in neurological autoimmune diseases similarly stress the need to evaluate both cellular and humoral vaccine responses under immunotherapy, reflecting the same integrated logic.

For neurologists, the case underscores a division of labor that can fail in fragmented health systems. Disease-modifying therapy is typically prescribed and monitored by a neurologist, while vaccinations may be administered by general practitioners, and acute infections present first to emergency departments. When no single actor owns the whole picture, vaccination assessment can fall through the cracks, and abnormal blood counts can be reviewed without connecting them to infection risk. The authors’ experience suggests that a simple checklist at each follow-up visit, covering current lymphocyte trajectory and vaccination status against influenza and pneumococcal disease, could close many of these gaps. The intensive care perspective adds weight to the argument: the clinicians writing this report encountered the end result of the missed safeguards, not the manageable outpatient problem that earlier intervention might have preserved.

It is worth stating clearly what the case does and does not show. A single report cannot establish that dimethyl fumarate caused the severe outcome, nor can it quantify the absolute risk of invasive pneumococcal disease among patients on the drug, which appears low given the large numbers treated worldwide. Network meta-analytic evidence supports the overall efficacy and acceptable safety of immunomodulators for relapsing-remitting multiple sclerosis, and dimethyl fumarate remains an appropriate choice for many patients. The value of the report lies in its illustration of a known mechanism, viral-bacterial synergy in the lung, meeting a known vulnerability, drug-associated lymphopenia, in a setting where both were potentially modifiable. Rare severe infections are precisely the kind of outcome that routine precautions are designed to prevent, and their occurrence signals a need to audit whether those precautions were actually in place.

The broader takeaway for patients is equally direct. Anyone living with multiple sclerosis and taking an immunomodulatory therapy should know their vaccination record, ask whether influenza and pneumococcal vaccines are up to date, and confirm that blood counts are being checked on the recommended schedule. Anyone prescribing such therapy should treat the vaccination assessment and the lymphocyte count as inseparable parts of the same safety routine. The Belgian case ended in an intensive care unit, but its authors frame it not as a reason to abandon an effective drug rather as evidence that the infrastructure around the drug, the vaccines given and the blood drawn, matters as much as the prescription itself. In an era of increasingly powerful immunotherapies for neurological disease, that infrastructure is where preventable harm is most often caught, or missed.

Subject of Research: Severe pneumococcal pneumonia following influenza A in a multiple sclerosis patient treated with dimethyl fumarate

Article Title: Severe invasive pneumococcal pneumonia following influenza A in a patient with multiple sclerosis treated with dimethyl fumarate: the importance of vaccination assessment and hematologic monitoring

Article References: Severe invasive pneumococcal pneumonia following influenza A in a patient with multiple sclerosis treated with dimethyl fumarate: the importance of vaccination assessment and hematologic monitoring. (n.d.). https://doi.org/10.1007/s00415-026-14118-z

Image Credits: AI Generated

DOI: 10.1007/s00415-026-14118-z

Keywords: multiple sclerosis, dimethyl fumarate, influenza A, pneumococcal pneumonia, lymphopenia, vaccination, hematologic monitoring, immunotherapy, Streptococcus pneumoniae, acute respiratory distress syndrome, disease-modifying therapy, intensive care

The review, led by Mishaal Munir and Rupesh Kotecha of the Baptist Health Herbert Wertheim Cancer Institute in Miami, together with colleagues at Florida International University and NYU Grossman School of Medicine, followed the PRISMA 2020 reporting framework. The team searched PubMed for studies of adults with solid-tumor leptomeningeal disease treated with one of three radiation approaches: whole-brain radiotherapy or other non-craniospinal photon techniques, photon craniospinal irradiation, or proton craniospinal irradiation. From 106 initially identified records, the investigators screened 91 after removing duplicates, reviewed 79 full-text reports, and ultimately included 32 unique studies—one of which contributed separate proton and photon craniospinal arms, bringing the total to 33 modality-specific analytic contributions.

The headline finding concerns overall survival. The median of the reported median survival values was 4.5 months across the 21 whole-brain radiotherapy studies, 4.1 months across the 6 photon craniospinal irradiation studies, and 10.7 months across the 6 proton craniospinal irradiation studies. Study-level medians in the proton group ranged from 4.0 to 13.7 months, while the whole-brain group spanned 1.9 to 14.5 months and the photon craniospinal group clustered between 1.9 and 4.8 months. Progression-free survival, reported far less consistently, averaged 3.9 months in the whole-brain group and 4.6 months in the proton group, with no extractable value for photon craniospinal treatment.

Why would protons make such a difference? The answer lies in fundamental radiation physics. Conventional photon beams, whether generated by linear accelerators for whole-brain or craniospinal treatment, deposit dose along their entire path and continue through the body on the exit side. When the target is the entire craniospinal axis—brain, spinal canal, and the cerebrospinal fluid pathways where leptomeningeal disease spreads—photons inevitably irradiate vertebral bone marrow, the heart, lungs, esophagus, and bowel on entry and exit. Proton beams, by contrast, can be shaped to release most of their energy at a precise depth, the Bragg peak, dramatically reducing exit dose to anterior organs and potentially sparing the marrow reservoirs that produce blood cells.

That dosimetric advantage shows up clearly in the toxicity data, though the authors caution that reporting was incomplete and heterogeneous. Among studies with extractable denominators, crude pooled rates of grade 3 or higher hematologic toxicity were 2.8 percent for whole-brain radiotherapy (4 of 145 patients), 80.0 percent for photon craniospinal irradiation (20 of 25), and 59.4 percent for proton craniospinal irradiation (38 of 64). Severe lymphopenia—the depletion of infection-fighting lymphocytes, which is increasingly recognized as a driver of poor outcomes in brain radiotherapy—dominated the craniospinal figures, affecting 80.0 percent of evaluable photon craniospinal patients and 58.6 percent of proton patients, versus 26.7 percent of evaluable whole-brain patients. Severe anemia, neutropenia, and thrombocytopenia were less common but more frequent in the photon craniospinal group.

The review’s context matters enormously here. The proton craniospinal literature is younger and, by design, more selective: these studies generally enrolled fitter patients treated in the modern era, often alongside CNS-active systemic therapies such as targeted agents and intrathecal treatments. The photon craniospinal studies, by contrast, largely date from older treatment periods with less conformal techniques and more protracted fractionation schedules. The authors are explicit that the numerical survival gap does not establish proton superiority over photon craniospinal irradiation, and they frame all cross-modality comparisons as descriptive and hypothesis-generating rather than estimates of comparative efficacy.

Nevertheless, the findings align with a striking randomized signal. In a phase II trial by Jonathan Yang and colleagues, proton craniospinal irradiation improved central nervous system progression-free survival to 8.2 months versus 2.3 months, and overall survival to 11.3 months versus 4.9 months, compared with photon involved-field radiotherapy in patients with breast cancer or non-small cell lung cancer leptomeningeal metastasis—without an increase in serious treatment-related adverse events. The hazard ratio for overall survival in multivariable analysis was 0.50. Yet the comparator in that trial was involved-field radiotherapy, not modern photon craniospinal treatment, leaving the proton-versus-photon craniospinal question formally unresolved. A real-world cohort from Frechette and colleagues reported a median survival of just 3.0 months with proton craniospinal irradiation, underscoring how powerfully patient selection shapes outcomes in this disease.

The whole-brain radiotherapy literature, the largest and most heterogeneous body of evidence, reflects decades of palliative practice. Whole-brain treatment can relieve intracranial symptoms and address coexisting brain metastases, but it does not cover the spinal cerebrospinal fluid compartment where leptomeningeal disease so often progresses. The 4.5-month median survival observed in this review is consistent with the historically grim prognosis documented in prior syntheses. Notably, one study excluded from quantitative pooling—Perlow and colleagues’ vertebral body–sparing photon craniospinal technique—reported no grade 3 or higher toxicity, hinting that hematologic toxicity may be as much a matter of technique as of particle type.

What comes next could settle the debate. The phase III NRG-BN014/RADIATE-LM trial (NCT06500481) is evaluating proton craniospinal irradiation against involved-field radiotherapy in breast cancer and non-small cell lung cancer leptomeningeal metastasis, though its comparator again leaves proton versus modern photon craniospinal irradiation unanswered. The review’s limitations are real: only PubMed was searched, no formal risk-of-bias scoring was applied, most included studies were retrospective, and progression endpoints were inconsistently defined. Still, by placing three radiation modalities in a single descriptive framework and pairing survival with toxicity, the analysis gives oncologists a benchmark for a decision that increasingly demands balancing potential disease control against treatment burden—and it signals that for a disease once considered uniformly terminal, the technology used to deliver radiation may be reshaping what survival looks like.

Subject of Research: Radiotherapy modalities and outcomes in solid-tumor leptomeningeal disease

Article Title: Survival and hematologic toxicity after whole-brain radiotherapy, photon craniospinal irradiation, and proton craniospinal irradiation for solid-tumor leptomeningeal disease: a systematic review

Article References: Survival and hematologic toxicity after whole-brain radiotherapy, photon craniospinal irradiation, and proton craniospinal irradiation for solid-tumor leptomeningeal disease: a systematic review. (n.d.). https://doi.org/10.1007/s11060-026-05776-0

Image Credits: AI Generated

DOI: 10.1007/s11060-026-05776-0

Keywords: leptomeningeal disease, proton craniospinal irradiation, photon craniospinal irradiation, whole-brain radiotherapy, systematic review, overall survival, hematologic toxicity, lymphopenia, brain metastases, radiation oncology, breast cancer, non-small cell lung cancer

Lymphocytes, the white blood cells that include T cells, B cells, and natural killer cells, are the sentinels of the adaptive immune system. They are also among the most radiosensitive cells in the human body. Unlike most tissues, however, lymphocytes do not sit still in one place during a course of radiotherapy. They circulate continuously through the bloodstream, which means that during brain irradiation, a substantial fraction of the body’s entire lymphocyte pool passes through the irradiated volume and is destroyed as it transits the treatment field. This mechanism, first modeled in detail by researchers at Johns Hopkins more than a decade ago, explains why even highly conformal modern radiotherapy can inflict significant collateral damage on the immune system.

The clinical stakes are considerable. Multiple studies, including a 2024 systematic review and pooled analysis, have found that patients who develop severe lymphopenia during chemoradiotherapy for glioblastoma survive substantially shorter periods than those who maintain their lymphocyte counts. Some investigators have proposed that radiation-induced immune depletion may blunt any potential benefit of immunotherapy in these patients and may even impair the body’s intrinsic ability to control residual tumor cells. What has remained contentious is precisely which dosimetric parameters best predict the severity of lymphocyte loss, and whether the culprit is the volume of brain irradiated, the dose delivered, or simply the size of the tumor target itself.

To address this question, Paolo Tini and colleagues conducted a retrospective single-center cohort study of 102 adults with histologically confirmed IDH-wildtype glioblastoma, the most common and aggressive molecular subtype of the disease. All patients had paired blood counts taken before radiotherapy and at treatment completion, along with detailed dose-volume histogram data for the whole brain and precise measurements of their target volumes—the gross tumor volume, clinical target volume, and planning target volume. The researchers also accounted for a battery of potential confounders, including baseline lymphocyte count, corticosteroid exposure, temozolomide administration, infection, and the number of radiation fractions delivered.

The results were striking. Median absolute lymphocyte counts fell from 1,810 cells per microliter before treatment to 1,050 cells per microliter at the end of radiotherapy, a median relative decline of 42.6 percent. Clinically significant lymphopenia was common: 32.4 percent of patients developed grade 2 or worse lymphopenia according to the Common Terminology Criteria for Adverse Events, and 13.7 percent developed grade 3 or worse. These figures underscore that immune depletion is not a rare toxicity but a near-universal feature of standard glioblastoma chemoradiotherapy, occurring in roughly one in three patients at a severity level that clinicians would typically flag as clinically meaningful.

The dosimetric analysis revealed a coherent and consistent pattern. In multivariable robust linear regression models adjusted for baseline lymphocyte count and major confounders, several measures of cerebral dose burden independently predicted lower lymphocyte counts at the end of treatment. Mean brain dose showed a significant association, with each additional 5 Gy reducing the log-transformed lymphocyte count by a factor corresponding to a beta coefficient of −0.251. The volume of brain receiving 15, 20, and 25 Gy—often abbreviated V15Gy, V20Gy, and V25Gy—each showed significant associations, as did the integral dose, a metric that captures the total energy deposited in the brain. Notably, V10Gy was also significant, while V5Gy was not, suggesting that intermediate-dose exposure in the range of roughly 15 to 25 Gy represents the most consistent dosimetric signal.

Perhaps the most provocative finding concerns the relationship between target volume and lymphocyte depletion. The planning target volume was strongly correlated with V20Gy, with a Spearman correlation coefficient of 0.78, meaning that larger targets inevitably expose more brain to intermediate doses. Yet when the researchers examined the direct association between target volume and lymphocyte preservation, it was negligible—a correlation of just −0.06, with a p-value of 0.55 indicating no statistically meaningful relationship. In other words, target size alone does not explain the immune damage. Two patients with similarly sized tumors could experience very different degrees of lymphocyte depletion depending on how the radiation dose is distributed through the surrounding brain, a conclusion that carries immediate implications for treatment planning.

The prognostic implications were equally clear. Using a landmark analysis anchored at the completion of radiotherapy—a design that avoids the statistical pitfalls of immortal time bias—the researchers found that grade 2 or worse lymphopenia was independently associated with inferior overall survival, with a hazard ratio of 2.43, and inferior progression-free survival, with a hazard ratio of 2.42. Patients whose immune systems were depleted during treatment faced more than double the risk of death and disease progression compared with those who preserved their lymphocyte counts, even after adjustment for other clinical and molecular variables. While the retrospective design of the study means that causality cannot be definitively established—lymphopenia could theoretically be a marker of some other adverse factor—the consistency of the association across multiple independent cohorts lends considerable weight to the biological interpretation.

The findings arrive at a moment when the field of radiation oncology is increasingly attentive to the immunological consequences of treatment. Prior work has shown that reducing the radiation treatment volume can lessen lymphopenia, and a randomized phase II study has demonstrated that proton therapy, which deposits less exit dose in normal tissue, reduces the likelihood of high-grade radiation-induced lymphopenia compared with conventional photon therapy. Other investigations have explored the role of intensity-modulated radiotherapy techniques and even the dose delivered to brain vasculature through which lymphocytes traffic. The Siena study adds an important nuance to this conversation: it is not enough to shrink the target or switch particle types. Planners must actively attend to the dose bath delivered to the otherwise healthy brain, particularly the intermediate-dose region between 15 and 25 Gy.

The authors argue that their findings support the prospective evaluation of immune-sparing radiotherapy strategies that reduce unnecessary cerebral dose exposure without compromising coverage of the tumor target. Such strategies might include tighter conformality, dose-escalation constraints on the normal brain, proton therapy in selected cases, or modified fractionation schedules. Given that glioblastoma remains one of the deadliest human cancers, with median survival measured in months despite aggressive multimodal treatment, any intervention that could preserve immune function without sacrificing tumor control deserves rigorous testing. This study, conducted without external funding and approved by the local ethics committee, offers a compelling dosimetric roadmap for that effort—and a reminder that in the quest to kill tumor cells, the collateral cost to the immune system may be a modifiable determinant of how long patients live.

Subject of Research: The association between cerebral radiation dose burden, treatment-related lymphopenia, and survival in IDH-wildtype glioblastoma

Article Title: Brain dose burden is associated with treatment-related lymphopenia in IDH-wildtype glioblastoma: dosimetric determinants and prognostic impact

Article References: Tini, P., Donnini, F., Battaglia, G., Pastina, P., Rubino, G., Carfagno, T., Vannini, M., Didona, A., Vanzi, E., & Banci Buonamici, F. (2026). Brain dose burden is associated with treatment-related lymphopenia in IDH-wildtype glioblastoma: dosimetric determinants and prognostic impact. Journal of Neuro-Oncology, 179(2), Article 81. https://doi.org/10.1007/s11060-026-05801-2

Image Credits: AI Generated

DOI: 10.1007/s11060-026-05801-2

Keywords: glioblastoma, lymphopenia, radiotherapy, brain dosimetry, V20Gy, integral dose, mean brain dose, temozolomide, immune-sparing radiotherapy, proton therapy, overall survival, IDH-wildtype

DRESS/DIHS is one of the most serious delayed-type adverse drug reactions encountered in clinical medicine. It is characterized by a widespread skin eruption, fever, systemic inflammation, involvement of internal organs such as the liver, hematologic abnormalities including atypical lymphocytes, and, in many cases, reactivation of human herpesvirus 6, or HHV-6. The syndrome typically develops two to six weeks after exposure to the culprit drug, a latency that distinguishes it from immediate hypersensitivity reactions. Mortality is significant, particularly when liver involvement is severe, and recognition depends on scoring systems such as RegiSCAR that integrate rash morphology, fever, lymphadenopathy, blood count abnormalities, organ involvement, and viral reactivation. In immunocompetent patients, the diagnostic constellation is usually recognizable, but the syndrome’s presentation can be substantially altered in patients whose immune systems have been profoundly suppressed by chemotherapy.

CPX-351, marketed as Vyxeos, is a dual-drug liposomal encapsulation designed to optimize the synergistic molar ratio of cytarabine to daunorubicin and to prolong their exposure within the plasma. It is approved for treatment-related acute myeloid leukemia and acute myeloid leukemia with myelodysplasia-related changes, populations in which intensified traditional induction regimens carry prohibitive toxicity. Skin eruptions are common adverse events with CPX-351, and one of the most frequent cutaneous findings in these patients is toxic erythema of chemotherapy, or TEC, a benign and self-limited eruption that typically affects the hands, intertriginous areas, and trunk within days of drug exposure. Distinguishing TEC from an evolving DRESS/DIHS-like reaction is difficult, and the new case illustrates precisely how the two conditions can overlap in their earliest phases.

In the reported patient, the cutaneous story began with an early eruption localized to the intertriginous areas and trunk, a distribution and timing initially compatible with toxic erythema of chemotherapy. Had the eruption followed the usual course, it would have been expected to stabilize and resolve without intervention. Instead, the clinical picture deteriorated. The patient developed recurrent fevers, and the eruption progressed to generalized erythema covering more than 90 percent of his body surface area. Accompanying features included lymphadenopathy, the appearance of atypical lymphocytes in the peripheral blood, and elevation of liver enzymes, each of which is a recognized component of the DRESS/DIHS diagnostic framework. Taken together, these findings shifted the clinical suspicion away from benign toxic erythema and toward a systemic drug hypersensitivity syndrome.

Several laboratory investigations provided pivotal diagnostic clues. Most strikingly, quantitative polymerase chain reaction demonstrated a markedly elevated HHV-6 DNA level of 1.2 million copies per milliliter of blood. HHV-6 reactivation is considered a signature feature of classic DRESS/DIHS, occurring in the majority of typical cases, and its detection in this severely lymphopenic patient served as a strong corroborating marker of the syndrome. At the same time, the case defied several conventional expectations. The patient had severe chemotherapy-induced lymphopenia, there was no peripheral eosinophilia, skin histology revealed inconspicuous dermal eosinophils, and serum thymus and activation-regulated chemokine, known as TARC and often used as a biomarker of DRESS/DIHS, was at a normal level. The authors emphasize that these deviations are precisely what made the diagnosis so difficult, since eosinophilia and elevated TARC are among the criteria clinicians most commonly rely upon.

A skin biopsy provided additional support for the diagnosis. Histopathologic examination showed interface dermatitis with superficial perivascular lymphocytic infiltration, a pattern consistent with a delayed hypersensitivity reaction rather than the necrotic keratinocytes and eccrine involvement more characteristic of toxic erythema of chemotherapy. Interface dermatitis, in which lymphocytes attack the basal layer of the epidermis, is a histologic hallmark seen in many drug-induced eruptions, including DRESS/DIHS, and its identification helped consolidate the clinical and laboratory findings into a coherent diagnosis despite the absence of eosinophils in the dermal infiltrate.

Treatment with systemic prednisolone produced a rapid and convincing response. The patient’s fever resolved promptly, appetite improved, the extensive skin lesions began to heal, atypical lymphocytes disappeared from the circulation, and HHV-6 DNA became undetectable. The temporal relationship between corticosteroid initiation and the simultaneous resolution of inflammatory, cutaneous, and virologic abnormalities is characteristic of DRESS/DIHS and further reinforced the diagnostic conclusion. Importantly, the patient was subsequently able to proceed to conditioning therapy and umbilical cord blood transplantation, meaning that recognition and treatment of the hypersensitivity reaction did not derail his definitive leukemia management.

Attribution of the reaction to CPX-351 required careful weighing of alternatives, and the authors frame the association as possible rather than definitive. The temporal relationship between CPX-351 administration and the onset of the reaction was the principal basis for implicating the drug, but the authors explicitly note that toxic erythema of chemotherapy and concomitant medications represented important alternative considerations. Patients with acute myeloid leukemia typically receive multiple supportive agents, including antimicrobials, antifungals, and antivirals, several of which are themselves recognized causes of DRESS/DIHS. In addition, the leukemia itself and its complications can produce fever, rash, hepatic dysfunction, and lymphadenopathy. The heterogeneity of the clinical environment in hematology patients means that drug causality assessment in this population is intrinsically uncertain, and the report’s cautious language reflects this reality.

Beyond its immediate clinical message, the case carries broader implications for how DRESS/DIHS is understood and diagnosed. Classical descriptions assume a relatively intact immune system capable of mounting the full hypersensitivity phenotype, including expansion of drug-specific T cells, eosinophil recruitment through interleukin-5, and elevated TARC production by activated lymphocytes. A patient with profound chemotherapy-induced lymphopenia lacks the cellular substrate for many of these features, and the reported case demonstrates that the syndrome can nonetheless develop in this setting. This suggests that current diagnostic criteria, which award points for eosinophilia, atypical lymphocytes, and lymphadenopathy, may underdiagnose DRESS/DIHS in severely immunocompromised patients. The authors propose that HHV-6 reactivation may serve as a particularly valuable diagnostic marker when eosinophilia is absent, since viral reactivation appears to depend on mechanisms that can persist despite lymphocyte depletion.

The report also offers practical guidance for clinicians who treat high-risk acute myeloid leukemia with CPX-351. Skin eruptions during therapy should not automatically be dismissed as toxic erythema of chemotherapy, particularly when they progress beyond the typical distribution, persist or worsen after the expected time course, or are accompanied by recurrent fever, lymphadenopathy, hepatic enzyme elevation, or atypical lymphocytes. In such circumstances, measurement of HHV-6 DNA and consideration of a DRESS/DIHS-like reaction are warranted even in the face of severe lymphopenia and absent eosinophilia. Early recognition matters because untreated DRESS/DIHS can progress to fulminant hepatic failure and death, while timely systemic corticosteroids, as demonstrated in this patient, can produce rapid clinical improvement and allow the underlying leukemia treatment plan, including allogeneic transplantation, to continue on schedule. As CPX-351 use expands worldwide, the authors suggest that clinicians maintain heightened awareness that this liposomal chemotherapy regimen, like many other drugs, may in rare cases trigger systemic hypersensitivity, and that its cutaneous and systemic fingerprints may look different in the immunosuppressed host than in the patients in which the syndrome was first described.

Subject of Research: Possible CPX-351-associated DRESS/DIHS-like drug hypersensitivity reaction during severe lymphopenia in acute erythroid leukemia

Article Title: Possible CPX-351-associated drug reaction with eosinophilia and systemic symptoms/drug-induced hypersensitivity syndrome-like reaction during severe lymphopenia

Article References: Bamba, S., Kanda, J., Hada, M., Kubota, Y., Nakajima, S., Nomura, T., Fukutome, T., Tsuji, K., & Takaori-Kondo, A. (2026). Possible CPX-351-associated drug reaction with eosinophilia and systemic symptoms/drug-induced hypersensitivity syndrome-like reaction during severe lymphopenia. Annals of Hematology. https://doi.org/10.1007/s00277-026-07282-9

Image Credits: AI Generated

DOI: 10.1007/s00277-026-07282-9

Keywords: CPX-351, DRESS, DIHS, HHV-6, acute myeloid leukemia, drug hypersensitivity, lymphopenia, skin eruptions, TARC, interface dermatitis, toxic erythema of chemotherapy, Annals of Hematology